def test_client_can_make_application_requests_without_api_key(dummy_data):
stub(responses.POST, "https://api.nexmo.com/v1/calls")
client = vonage.Client(application_id="myid",
private_key=dummy_data.private_key)
voice = vonage.Voice(client)
voice.create_call("123455")
def call(self):
voice = vonage.Voice(self.client)
text = 'Hi there, domain {}.{} is available now.'.format(self.domain, self.tld)
response = voice.create_call({
'to': [{'type': 'phone', 'number': "6587837006"}],
'from': {'type': 'phone', 'number': "6587837006"},
'ncco': [{'action': 'talk', 'text': text}]
})
if (response["status"] == "started"):
print("| Reported to Son dai ka ^o^ ")
else :
print("| Could not contact Son dai ka @.@")
def test_user_provided_authorization(client, dummy_data):
stub(responses.GET, "https://api.nexmo.com/v1/calls/xx-xx-xx-xx")
application_id = "different-nexmo-application-id"
nbf = int(time.time())
exp = nbf + 3600
client.auth(application_id=application_id, nbf=nbf, exp=exp)
voice = vonage.Voice(client)
voice.get_call("xx-xx-xx-xx")
token = request_authorization().split()[1]
token = jwt.decode(token, dummy_data.public_key, algorithms="RS256")
assert token["application_id"] == application_id
assert token["nbf"] == nbf
assert token["exp"] == exp
def test_authorization_with_private_key_path(dummy_data):
stub(responses.GET, "https://api.nexmo.com/v1/calls/xx-xx-xx-xx")
private_key = os.path.join(os.path.dirname(__file__),
"data/private_key.txt")
client = vonage.Client(
key=dummy_data.api_key,
secret=dummy_data.api_secret,
application_id=dummy_data.application_id,
private_key=private_key,
)
voice = vonage.Voice(client)
voice.get_call("xx-xx-xx-xx")
token = jwt.decode(request_authorization().split()[1],
dummy_data.public_key,
algorithms="RS256")
assert token["application_id"] == dummy_data.application_id
dotenv_path = join(dirname(__file__), "../.env")
load_dotenv(dotenv_path)
VONAGE_APPLICATION_ID = os.environ.get("VONAGE_APPLICATION_ID")
VONAGE_APPLICATION_PRIVATE_KEY_PATH = os.environ.get(
"VONAGE_APPLICATION_PRIVATE_KEY_PATH")
VONAGE_NUMBER = os.environ.get("VONAGE_NUMBER")
TO_NUMBER = os.environ.get("TO_NUMBER")
client = vonage.Client(
application_id=VONAGE_APPLICATION_ID,
private_key=VONAGE_APPLICATION_PRIVATE_KEY_PATH,
)
voice = vonage.Voice(client)
response = voice.create_call({
'to': [{
'type': 'phone',
'number': TO_NUMBER
}],
'from': {
'type': 'phone',
'number': VONAGE_NUMBER
},
'ncco': [{
'action': 'talk',
'text': 'This is a text to speech call from Nexmo'
}]
})
def voice(client, dummy_data):
import vonage
return vonage.Voice(client)
#!/usr/bin/env python3
import vonage, os
from pprint import pprint
from datetime import datetime, timedelta
from os.path import join, dirname
from dotenv import load_dotenv
dotenv_path = join(dirname(__file__), "../.env")
load_dotenv(dotenv_path)
VONAGE_APPLICATION_ID = os.environ.get("VONAGE_APPLICATION_ID")
VONAGE_APPLICATION_PRIVATE_KEY_PATH = os.environ.get(
"VONAGE_APPLICATION_PRIVATE_KEY_PATH")
voice = vonage.Voice(
application_id=VONAGE_APPLICATION_ID,
private_key=VONAGE_APPLICATION_PRIVATE_KEY_PATH,
)
NOW = datetime.utcnow()
DATE_END = NOW.replace(microsecond=0).isoformat() + "Z"
DATE_START = (NOW - timedelta(hours=24, minutes=00)).replace(
microsecond=0).isoformat() + "Z"
response = voice.get_calls(date_start=DATE_START, date_end=DATE_END)
calls = response['_embedded']['calls']
for call in calls:
pprint(call)
def voice(self):
ctx = _app_ctx_stack.top
if ctx is not None:
if not hasattr(ctx, 'vonage_voice'):
ctx.vonage_voice = vonage.Voice(self.client)
return ctx.vonage_voice
def respond(voice_data):
# 0: (Vonage) Initiate Video Meeting (1-on-1)
if there_exists(["video", "meeting", "conference"]):
dest_phone_number = record_audio(
f"What is the phone number of the person you seek to invite?")
confirm = record_audio(
f"The phone number you said was {dest_phone_number}, correct?")
attempt = 0 # Initializes variable (int)
while (True):
if confirm == "yes" or confirm == "yes ma'am" or confirm == "correct" or confirm == "affirmative" or confirm == "confirmed" or confirm == "right" or confirm == "yup" or confirm == "yep": # User confirmed the desired recipient's phone number
dest_phone_number_int = int(
str(1) + dest_phone_number.replace('-', '')
) # From variable (string), remove dashes (-), prepend US country code (1), and convert to int (i.e. 123-456-7890 --> 1234567890) to match Vonage's "Send SMS" API field "to:" requirement. This assumes the user didn't provide an incongruent US-based phone number, nor a US country code)
client = vonage.Client(
key=martha_vonage_api_key, secret=martha_vonage_api_secret
) # Sets client (Martha's) parameters for Vonage account
response = vonage.Sms(client).send_message({ # Sends SMS message and stores JSON response (useful for potential logging and troubleshooting)
"from": martha_vonage_phone,
"to": dest_phone_number_int,
"text": "Hi. You have been invited to join the following video meeting: " + martha_vonage_video_url + " .",
})
webbrowser.get().open(
martha_vonage_video_url) # Open video meeting URL
speak(
f"Your video meeting invitation has been sent to 1-{dest_phone_number}. I will leave you now to meet with them in private. Sayonara {person_obj.name}"
)
print(
f"Martha: Video Meeting Link: {martha_vonage_video_url}"
) # Print (to stdout) the video meeting URL, in case user mistakenly closes their browser tab prematurely and wishes to return to the meeting
exit() # Close Martha application
else: # User did not confirm the desired recipient's phone number
if attempt == 2: # "Initiate Video Meeting" request aborts, after 3 tries
speak(
f"I'm afraid I don't understand. What else might I be able to help you with today {person_obj.name}?"
)
break
else: # User asked to provide recipient's phone number again
dest_phone_number = record_audio(
f"I'm sorry. Please repeat the recipient's phone number."
)
confirm = record_audio(
f"The phone number you said was {dest_phone_number}, correct?"
)
attempt = attempt + 1
continue
# 1: (Vonage) Send SMS
elif there_exists(["text", "message", "sms"]):
dest_phone_number = voice_data.split("text ")[-1]
dest_phone_number = dest_phone_number.split("message ")[-1]
dest_phone_number = dest_phone_number.split("sms")[-1]
dest_phone_number = dest_phone_number.split("to ")[-1]
confirm = record_audio(
f"The phone number you said was {dest_phone_number}, correct?"
) # Technically, this section should perform input validation (i.e. handling of phone number lengths and country codes). A lengthy commented out code block follows further down that takes an initial stab at this, but I dropped the effort, because I didn’t want to get mired into input validation edge case logic during my initial development
attempt = 0 # Initializes variable (int)
while (True):
if confirm == "yes" or confirm == "yes ma'am" or confirm == "correct" or confirm == "affirmative" or confirm == "confirmed" or confirm == "right" or confirm == "yup" or confirm == "yep": # User confirmed the desired recipient's phone number
text_message = record_audio(
f"What would you like to text the recipient?")
confirm = record_audio(
f"You would like to text: \"{text_message}\". Is that correct {person_obj.name}?"
)
attempt = 0 # Initializes variable (int)
while (True):
if confirm == "yes" or confirm == "yes ma'am" or confirm == "correct" or confirm == "affirmative" or confirm == "confirmed" or confirm == "right" or confirm == "yup" or confirm == "yep": # User confirmed the desired message to be sent
dest_phone_number_int = int(
str(1) + dest_phone_number.replace('-', '')
) # From variable (string), remove dashes (-), prepend US country code (1), and convert to int (i.e. 123-456-7890 --> 1234567890) to match Vonage's "Send SMS" API field "to:" requirement. This assumes the user didn't provide an incongruent US-based phone number, nor a US country code)
"""
# This commented out code block performs phone number input validation (i.e. handling of phone numbers and country codes). It works, but is incomplete (i.e. short codes). It’s also a lot more complicated than the single line of code above, which is why it wasn’t utilized. I’ve left it here for anyone desiring to continue this software development effort however, as input validation is important when it comes to production worthy code. Also, it should technically be moved into (and before) the first while loop (i.e. when validating the phone number provided)
dest_phone_number_str = dest_phone_number.replace('-', '').replace(' ','') # From variable (string), remove dashes (-) and spaces ( ). For instance, this will convert "123-456-7890" to "1234567890"
if dest_phone_number_str[0] == 1 and len(dest_phone_number_str) == 11: # Simply convert phone number to an int, given phone number appears to of appropriate length and possesses a US country code (+1)
dest_phone_number_int = int(dest_phone_number_str) # Int required to match Vonage's "Send SMS" API field "to:" requirement
speak(f"Scenario A")
elif dest_phone_number_str[0] != 1 and len(dest_phone_number_str) == 10: # Add US country code (+1) and convert to an int, given phone number is the right length, except for the missing US area code (+1). This scenario is the one most likely to execute
dest_phone_number_int = int(str(1) + dest_phone_number_str)
speak(f"Scenario B")
else: # Phone number provided is incogruent with US phone standards, is a short code, mistakenly includes an extension, and other potential catch-alls
speak(f"Please provide an appropriate US based 10 digit length phone number, without extensions. You do not need to specify the US country code.")
speak(f"Scenario C")
break
"""
client = vonage.Client(
key=martha_vonage_api_key,
secret=martha_vonage_api_secret
) # Sets client (Martha's) parameters for Vonage account
response = vonage.Sms(client).send_message({ # Sends SMS message and stores JSON response (useful for potential logging and troubleshooting)
"from": martha_vonage_phone,
"to": dest_phone_number_int,
"text": text_message,
})
speak(
f"Your text \"{text_message}\" has been sent to 1-{dest_phone_number}. What else might I be able to help you with today {person_obj.name}?"
)
break
else: # User did not confirm the desired message to be sent
if attempt == 2: # "Send an SMS" request aborts, after 3 tries
speak(
f"I'm afraid I don't understand. What else might I be able to help you with today {person_obj.name}?"
)
break
else: # User asked to provide message to be sent again
text_message = record_audio(
f"I'm sorry. Please repeat the message you would like sent."
)
confirm = record_audio(
f"You would like to text: \"{text_message}\". Is that correct {person_obj.name}?"
)
attempt = attempt + 1
continue
break
else: # User did not confirm the desired recipient's phone number
if attempt == 2: # "Send an SMS" request aborts, after 3 tries
speak(
f"I'm afraid I don't understand. What else might I be able to help you with today {person_obj.name}?"
)
break
else: # User asked to provide recipient's phone number again
dest_phone_number = record_audio(
f"I'm sorry. Please repeat the recipient's phone number."
)
confirm = record_audio(
f"The phone number you said was {dest_phone_number}, correct?"
)
attempt = attempt + 1
continue
# 2: (Vonage) Running Late / Make (pre-scripted) Voice Call (text-to-speech via answer_url)
elif there_exists(["late", "tardy", "delayed", "running behind"]):
dest_phone_number = record_audio(
f"What is the recipient's phone number?")
confirm = record_audio(
f"The phone number you said was {dest_phone_number}, correct?"
) # Technically, this section should perform input validation (i.e. handling of phone number lengths and country codes). A lengthy commented out code block was provided in "(Vonage) Initiate Video Meeting (1-on-1)" that takes an initial stab at this, but I dropped the effort, because I didn’t want to get mired into input validation edge case logic during my initial development
attempt = 0 # Initializes variable (int)
while (True):
if confirm == "yes" or confirm == "yes ma'am" or confirm == "correct" or confirm == "affirmative" or confirm == "confirmed" or confirm == "right" or confirm == "yup" or confirm == "yep": # User confirmed the desired recipient's phone number
dest_phone_number_int = int(
str(1) + dest_phone_number.replace('-', '')
) # From variable (string), remove dashes (-), prepend US country code (1), and convert to int (i.e. 123-456-7890 --> 1234567890) to match Vonage's "Make Call" API field "to:" requirement. This assumes the user didn't provide an incongruent US-based phone number, nor a US country code)
client = vonage.Client(
application_id=martha_vonage_application_id,
private_key=martha_vonage_application_private_key_path)
voice = vonage.Voice(client)
response = voice.create_call({
'to': [{
'type': 'phone',
'number': dest_phone_number_int
}],
'from': {
'type': 'phone',
'number': martha_vonage_phone
},
'answer_url': [voice_running_late_answer_url]
})
speak(
f"I called to inform 1-{dest_phone_number} that you are running ten minutes late. What else might I be able to help you with today {person_obj.name}?"
)
break
else: # User did not confirm the desired recipient's phone number
if attempt == 2: # "Running Late" request aborts, after 3 tries
speak(
f"I'm afraid I don't understand. What else might I be able to help you with today {person_obj.name}?"
)
break
else: # User asked to provide recipient's phone number again
dest_phone_number = record_audio(
f"I'm sorry. Please repeat the recipient's phone number."
)
confirm = record_audio(
f"The phone number you said was {dest_phone_number}, correct?"
)
attempt = attempt + 1
continue
# 3: (Vonage) Make Voice Call (text-to-speech via NCCO)
elif there_exists(["ring", "call", "voice note"]):
dest_phone_number = voice_data.split("ring ")[-1]
dest_phone_number = dest_phone_number.split("call ")[-1]
dest_phone_number = dest_phone_number.split("voice note ")[-1]
dest_phone_number = dest_phone_number.split("to ")[-1]
dest_phone_number = dest_phone_number.split("for ")[-1]
dest_phone_number = dest_phone_number.split("at ")[-1]
confirm = record_audio(
f"The phone number you said was {dest_phone_number}, correct?"
) # Technically, this section should perform input validation (i.e. handling of phone number lengths and country codes). A lengthy commented out code block was provided in "(Vonage) Initiate Video Meeting (1-on-1)" that takes an initial stab at this, but I dropped the effort, because I didn’t want to get mired into input validation edge case logic during my initial development
attempt = 0 # Initializes variable (int)
while (True):
if confirm == "yes" or confirm == "yes ma'am" or confirm == "correct" or confirm == "affirmative" or confirm == "confirmed" or confirm == "right" or confirm == "yup" or confirm == "yep": # User confirmed the desired recipient's phone number
ncco_voice_message = record_audio(
f"What would you like to tell the recipient?")
confirm = record_audio(
f"You would like to tell them: \"{ncco_voice_message}\". Is that correct {person_obj.name}?"
)
attempt = 0 # Initializes variable (int)
while (True):
if confirm == "yes" or confirm == "yes ma'am" or confirm == "correct" or confirm == "affirmative" or confirm == "confirmed" or confirm == "right" or confirm == "yup" or confirm == "yep": # User confirmed the desired message to be sent
dest_phone_number_int = int(
str(1) + dest_phone_number.replace('-', '')
) # From variable (string), remove dashes (-), prepend US country code (1), and convert to int (i.e. 123-456-7890 --> 1234567890) to match Vonage's "Make Call" Voice API field "to:" requirement. This assumes the user didn't provide an incongruent US-based phone number, nor a US country code)
client = vonage.Client(
application_id=martha_vonage_application_id,
private_key=
martha_vonage_application_private_key_path)
voice = vonage.Voice(client)
response = voice.create_call({
"to": [{
'type': 'phone',
'number': dest_phone_number_int
}],
"from": {
'type': 'phone',
'number': martha_vonage_phone
},
"ncco": [{
"action": "talk",
"text": ncco_voice_message
}]
})
speak(
f"Your message \"{ncco_voice_message}\" has been sent to 1-{dest_phone_number}. What else might I be able to help you with today {person_obj.name}?"
)
break
else: # User did not confirm the desired message to be sent
if attempt == 2: # "Send an SMS" request aborts, after 3 tries
speak(
f"I'm afraid I don't understand. What else might I be able to help you with today {person_obj.name}?"
)
break
else: # User asked to provide message to be sent again
ncco_voice_message = record_audio(
f"I'm sorry. Please repeat the message you would like conveyed."
)
confirm = record_audio(
f"You would like to tell them: \"{ncco_voice_message}\". Is that correct {person_obj.name}?"
)
attempt = attempt + 1
continue
break
else: # User did not confirm the desired recipient's phone number
if attempt == 2: # "Send an SMS" request aborts, after 3 tries
speak(
f"I'm afraid I don't understand. What else might I be able to help you with today {person_obj.name}?"
)
break
else: # User asked to provide recipient's phone number again
dest_phone_number = record_audio(
f"I'm sorry. Please repeat the recipient's phone number."
)
confirm = record_audio(
f"The phone number you said was {dest_phone_number}, correct?"
)
attempt = attempt + 1
continue
# 4: Vonage Headquarters
elif there_exists(["vonage headquarters"]):
webbrowser.get().open(
"https://www.google.nl/maps/place/Vonage/@40.347517,-74.1901178,17z/data=!3m1!4b1!4m5!3m4!1s0x89c2329e087380b1:0x61f1c7579b67b5d6!8m2!3d40.3475129!4d-74.1879291"
)
speak(
"Vonage headquarters is located at 23 Main Street, Holmdel, New Jersey. Vonage can be reached at 732-528-2600. Operating hours are Monday through Friday, from 9 AM to Midnight. As well as Saturday and Sunday, from 9 AM to 8 PM."
)
# 5: Get Stock Price
elif there_exists(["price of", "value of"]):
search_term = voice_data.lower().split(
" of "
)[-1] # Split captures all words AFTER " of " (assumes user will ONLY speak the search query afterwards)
stocks = {
"vonage": "VG",
"gamestop": "GME",
"bitcoin": "BTC-USD",
"tesla": "TSLA",
"google": "GOOGL"
}
try:
stock = stocks[
search_term] # Check dictionary (stocks) for company (search_term). If found, store the company's (key's) value (stock ticker symbol)
stock = yf.Ticker(
stock
) # Queries Yahoo Finance API for stock ticker symbol (i.e. VG) and returns a JSON object containing associated data (i.e. market price, volume, business sector, and number of full time employees)
speak(
f'The price of {search_term} is {stock.info["regularMarketPrice"]} ({stock.info["currency"]})'
) # Extracts desired JSON data (i.e. regular market price and associated currency) for speech inclusion
except: # Error Condition: Yahoo Finance (yfinance) API is down. Or, the requested company stock (i.e. Starbucks' SBUX) isn't represented in the "stocks" dictionary above
speak('Sorry, I am unfamiliar with that stock')
# 6: Query Vonage
elif there_exists(["search", "query"]):
search_term = voice_data.split("for ")[-1]
url = search_term.replace(" ", "+")
url = f"https://vonage.com/search/?q={url}"
webbrowser.get().open(url)
speak(f'Here is what I found for {search_term} on the vonage website')
# 7: Find Location
elif there_exists(["show me", "find", "map of", "location of",
"where is"]):
# Below "split()" series is a best-effort (ordered) means of parsing spoken English; please don't try to intentionally confuse Martha!
location = voice_data.split("show me ")[-1]
location = location.split("find ")[-1]
location = location.split("map of ")[-1]
location = location.split("location of ")[-1]
location = location.split("where is ")[-1]
url = f"https://google.nl/maps/place/{location}"
webbrowser.get().open(url)
speak("This is where " + location + " is located")
# 8: Name Inquiry
elif there_exists(["your name"]):
if person_obj.name: # Check if string is NOT empty (i.e. user's name is known)
speak("my name is Martha")
else:
person_name = record_audio("My name is Martha. What's your name?")
person_name = person_name.split("is ")[-1]
person_obj.setName(
person_name) # Store user's name for future recall
if person_obj.name == 'martha': # Check if user's name is Martha
speak(f"Wow, I can't believe we share the same name, Martha!")
else:
speak(
f"Hi {person_name}, it's my pleasure to make your acquaintence"
)
# 9: My Name
elif there_exists(["my name is"]):
person_name = voice_data.split("is ")[-1]
person_obj.setName(person_name)
if person_obj.name == 'martha': # Oh, you think you're funny? You best watch out; Martha can be cheeky!
speak(
f"Oh, really? Because, my name is Martha. I think you're just joshing me, so I'm just going to call you Josh"
)
person_obj.setName('Josh')
else:
speak(f"Okay, I will remember your name, {person_name}")
# 10: Exit Application
elif there_exists(
["exit", "quit", "goodbye", "sayonara", "take care", "farewell"]):
speak(f"Sayonara {person_obj.name}")
exit() # Close Martha application
# 11: Courteous Greeting
elif there_exists(["how are you"]):
speak(f"I'm very well, thanks for asking {person_obj.name}")
# 12: Basic Greeting
elif there_exists(['hey', 'hi', 'hello', 'howdy', 'hola', 'Konnichiwa']):
greetings = [
f"Hey, how can I help you {person_obj.name}?",
f"Hey, what's up {person_obj.name}?",
f"I'm listening {person_obj.name}",
f"How can I help you {person_obj.name}?",
f"Hello {person_obj.name}"
]
greet = greetings[random.randint(
0,
len(greetings) - 1
)] # Randomly selects a greeting from the above list. If the user's name is known, Martha will address them by name, too
speak(greet)
def voice(client):
import vonage
return vonage.Voice(client)
